182 / 322
期貨未平倉量的資訊內涵及其交易活動之研究
182
Appendix B
The three volatility proxies are identified by ARIMA models with the following
parameter estimates:
Numbers in parentheses are t-statistics. All coefficients are significant at the 5% level.
None of the Ljung-Box Q-statistics for lag lengths 1-60 are significant at the 5% level.